adding model-info and tokenize to the TC RHOAIENG-3483 (#1281)

Signed-off-by: mwaykole <[email protected]>
red-hat-data-services · Mar 13, 2024 · 155cefa · 155cefa
1 parent 394bac5
commit 155cefa
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 1 deletion.
diff --git a/ods_ci/tests/Resources/Files/llm/model_expected_responses.json b/ods_ci/tests/Resources/Files/llm/model_expected_responses.json
@@ -38,7 +38,10 @@
                 "llama-2-13b-chat": {
                     "response_tokens": 20,
                     "response_text": "\n\nWater boils at 100 degrees Celsius or 212",
-                    "streamed_response_text": "{    'inputTokenCount': 10}{    'generatedTokenCount': 2,    'text': '\n'}{    'generatedTokenCount': 3,    'text': 'I a'}{    'generatedTokenCount': 4,    'text': 'm no'}{    'generatedTokenCount': 5,    'text': 't sur'}{    'generatedTokenCount': 6,    'text': 'e i'}{    'generatedTokenCount': 7,    'text': 'f thi'}{    'generatedTokenCount': 8,    'text': 's i'}{    'generatedTokenCount': 9,    'text': 's th'}{    'generatedTokenCount': 10,    'text': 'e righ'}{    'generatedTokenCount': 11,    'text': 't plac'}{    'generatedTokenCount': 12,    'text': 'e t'}{    'generatedTokenCount': 13,    'text': 'o as'}{    'generatedTokenCount': 14,    'text': 'k thi'}{    'generatedTokenCount': 15,    'text': 's questio'}{    'generatedTokenCount': 16,    'text': 'n'}{    'generatedTokenCount': 17,    'text': ', bu'}{    'generatedTokenCount': 18,    'text': 't '}{    'generatedTokenCount': 19,    'text': 'I a'}{    'generatedTokenCount': 20,    'text': 'm trying',    'stopReason': 'MAX_TOKENS'}"
+                    "streamed_response_text": "{    'inputTokenCount': 10}{    'generatedTokenCount': 2,    'text': '\n'}{    'generatedTokenCount': 3,    'text': 'I a'}{    'generatedTokenCount': 4,    'text': 'm no'}{    'generatedTokenCount': 5,    'text': 't sur'}{    'generatedTokenCount': 6,    'text': 'e i'}{    'generatedTokenCount': 7,    'text': 'f thi'}{    'generatedTokenCount': 8,    'text': 's i'}{    'generatedTokenCount': 9,    'text': 's th'}{    'generatedTokenCount': 10,    'text': 'e righ'}{    'generatedTokenCount': 11,    'text': 't plac'}{    'generatedTokenCount': 12,    'text': 'e t'}{    'generatedTokenCount': 13,    'text': 'o as'}{    'generatedTokenCount': 14,    'text': 'k thi'}{    'generatedTokenCount': 15,    'text': 's questio'}{    'generatedTokenCount': 16,    'text': 'n'}{    'generatedTokenCount': 17,    'text': ', bu'}{    'generatedTokenCount': 18,    'text': 't '}{    'generatedTokenCount': 19,    'text': 'I a'}{    'generatedTokenCount': 20,    'text': 'm trying',    'stopReason': 'MAX_TOKENS'}",
+                    "tgis-runtime": {
+                        "tokenize_response_text": "{'responses':[{'tokenCount':9,'tokens':['\\u003cs\\u003e','▁At','▁what','▁temperature','▁does','▁water','▁bo','il','?']}]}"
+                    }
                 }
             }
         },
@@ -172,6 +175,9 @@
         },
         "mpt-7b-instruct2":{
             "tgis-runtime": "{     'maxSequenceLength': 2048,     'maxNewTokens': 1024   }"
+        },
+        "llama-2-13b-chat-hf":{
+            "tgis-runtime": "{ 'maxSequenceLength': 4096,'maxNewTokens': 1024}"
         }
     }
 }
diff --git a/...ests/Tests/400__ods_dashboard/420__model_serving/LLMs/422__model_serving_llm_models.robot b/...ests/Tests/400__ods_dashboard/420__model_serving/LLMs/422__model_serving_llm_models.robot
@@ -345,6 +345,14 @@ Verify User Can Serve And Query A meta-llama/llama-2-13b-chat Model
     ...    inference_type=streaming    n_times=1    protocol=grpc
     ...    namespace=${test_namespace}    query_idx=0    validate_response=${FALSE}
     ...    port_forwarding=${use_port_forwarding}
+    Query Model Multiple Times    model_name=${model_name}    runtime=${TGIS_RUNTIME_NAME}
+    ...    inference_type=model-info    n_times=0
+    ...    namespace=${test_namespace}    validate_response=${TRUE}    string_check_only=${TRUE}
+    ...    port_forwarding=${use_port_forwarding}
+    Query Model Multiple Times    model_name=${model_name}    runtime=${TGIS_RUNTIME_NAME}
+    ...    inference_type=tokenize    n_times=0    query_idx=0
+    ...    namespace=${test_namespace}    validate_response=${TRUE}    string_check_only=${TRUE}
+    ...    port_forwarding=${use_port_forwarding}
     [Teardown]    Run Keywords
     ...    Clean Up Test Project    test_ns=${test_namespace}
     ...    isvc_names=${models_names}    wait_prj_deletion=${FALSE}